Non invasive evaluation of end systolic left ventricular elastance according to pressure-volume curve modeling during ejection in arterial hypertension
Résumé
Objective: End systolic left ventricular (LV) elastance (Ees) has been previously
calculated and validated invasively using LV pressure-volume (P-V) loop. Non
invasive studies have been proposed, but clinical application remains complex.
The aim of the present study was to 1) estimate Ees according to modeling of
LV P-V curve during ejection (“Ejection PV Curve’’ method) and validate our
method with existing published LV P-V loop data; 2) test clinical applicability to
detect non invasively a difference in Ees between normotensive and hypertensive
subjects.
Design and method: Based on P-V curve and a linear relationship between LV elastance
and time during ejection, we used a non linear least square method to fi t the systolic
pressure curve. We then computed slope and intercept of time varying elastance,
and calculated Ees as LV elastance at the end of ejection. As a validation, 22 P-V loops
obtained from previous invasive studies were digitized and analyzed with our method.
To test clinical applicability, P-V curve was obtained from 33 hypertensive and 32
normotensive subjects, using carotid tonometry and real time 3D echocardiography.
Results: A good univariate relationship (r2 = 0.92, p < 0.005) and a good limit of
agreement were found between previous invasive calculation of Ees and our new
proposed “Ejection P-V Curve’’ method. In addition, the clinical reproductibility
of our method was similar to that of another non invasive method proposed by
Chen et al. In hypertensive patients, the increase in arterial elastance (Ea) was
compensated by an increase in Ees without change in Ea/Ees (see Figure).
Conclusions: Ees can be estimated non invasively from modeling of P-V curve during
ejection. This approach was found to be reproducible and sensitive enough to detect an
expected difference in LV contractility in hypertensive patients. Due to its non invasive
nature, this methodology may have clinical implications in various disease states.